Drosophila has long been used as a model organism in the study of vision. The Drosophila medulla is the color vision processing center of the visual system. It is composed of over 60 different types of neurons. I hypothesize that each of these medulla neuron types is specified by the combinatorial expression of a set of transcription factors (TFs), "TF-code". This proposal focuses on determining the code of TFs that controls the generation and specification of a set of these neuronal sub-types. As a test case, I will first characterize the genetic roles of two TFs, homothorax (hth) and apterous (ap), in the specification of the medulla Mi1 neuron, which is the only medulla neuron type that can be identified based only on molecular markers. I will then screen a library of 700 Gal4 lines inserted into TF loci for lines that are expressed in medulla neurons (NP- Consortium of Japan). This collection covers ~80% of Drosophila TFs and by identifying the medulla neurons that express all of these lines I will be able to define a large part of the TF-code of medulla neurons. In Aim 2, I will use flow cytometry and deep sequencing to analyze the genetic basis of medulla neuronal diversity. I will use Gal4 lines identified in Aim1 in combination with flip-out cassettes and split lexA technology to label single medulla neuron types for analysis by cell sorting and deep sequencing. This analysis will provide quantitative data on genome wide-gene expression for each medulla neuron type analyzed. It will generate insights into the genetic differences between columnar and non-columnar neuron types, local and projection neuron types, and will further refine the TF-code for each neuron sub-types analyzed. Additionally, this analysis will determine the neurotransmitter/neurotransmitter-receptor usage and ion-channel expression for each cell type analyzed. Finally, I will compare gene expression profiles between individual medulla neuron types in order to identify candidate TFs for genetic analysis in Aim 3. In this last aim, I will use a combination of gain and loss-of- function for candidate TFs identified in Aim 2 for phenotypes related to cell-fate specification for medulla neuron sub-types for which I have deep sequence data for from Aim 2. I will also transfer data and tools developed in this proposal on to other researchers in the host laboratory working on behavioral and electrophysiological approaches to understanding color vision in Drosophila. The data generated in this study will define the TF code for the medulla neuron types and provide more general insight into the genetic basis of cellular diversity in neuronal systems. Finally, establishing a robust pipeline for genome-wide expression analysis by cell sorting and deep sequencing will enable others to use this approach in studying other neuronal systems and structures. PUBLIC HEALTH RELEVANCE: Understanding the genetic specification of different neuron types will enable neurological disorders and neurodegenerative diseases to be better understood and treated. The work presented in this proposal will significantly advance current understanding of genetic differences between neuron types in the fly visual system and will provide more general insight into the genetic basis of neuronal diversity in neuronal systems.